Aug 25, 2023 This Week in Cardiology Podcast

John M. Mandrola, MD


August 25, 2023

Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast, download the Medscape app or subscribe on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only.

In This Week’s Podcast

For the week ending August 25, 2023 John Mandrola, MD comments on the following news and features stories.

Listener Feedback

I am broadcasting from the European Society of Cardiology (ESC) meeting in Amsterdam. First of all, I love this city. I hired a bike and am using it for transportation. And I look forward to seeing familiar faces and meeting new colleagues. Please say hello.

First, some listener feedback. Dr. Doron Zahger rightly pushed back on one of my comments last week. In making the argument that implantable cardioverter defibrillator (ICD) therapy likely has no substantial effect on mortality reduction in patients with nonischemic cardiomyopathy, I cited the subgroup analysis of SCD-HeFT. I took the view that most of the average benefit in SCD-HeFT concentrated in the ischemic cohort. Dr Zahger rightly points out a) that trials are powered for the overall population, and we should be very cautious making claims from subgroups; and b) that in the ICD arm of the SCD-HeFT subgroup in non-ischemics, the P-value was 0.06.

He is right, and as I was going through that discussion, my imaginary friend in the back of my brain, was saying be careful with that SCD-HeFT bullet point. I should have spoken more clearly. Thanks for listening so intently and taking the time to write.

TWIC podcast comes out every Friday and due to the time delay, I can cover some of the first trials presented at ESC. There won’t be much time for commentary or deep thinking on them, but I can offer the topline results and some quick thoughts.


The STEP-HFpEF trial combines three of the biggest topics in Medicine today: GLP-1 agonists, obesity, and heart failure with preserved ejection fraction (HFpEF). HFpEF is the most common cause of HF. Obesity is surely a condition that worsens quality of life in patients with HFpEF. Obesity is also a likely cause of HFpEF. The GLP-1 agonists have clearly been shown to induce serious weight loss in patients with obesity.

STEP-HFpEF combines these two givens. That is, if the drug induces weight loss, it ought to improve functional capacity and quality of life.

  • The trial enrolled about 500 patients with HFpEF and obesity.

  • One group was randomly assigned to semaglutide, the other placebo.

  • Patients had to have a body mass index (BMI) > 30 kg/m2 and an EF ≥ 45%.

  • HFpEF meant having an elevated filling pressure at catheterization, or an implantable monitor, or an elevated NT-proBNP with typical echocardiographic features of HFpEF, or a hospitalization for HF requiring IV diuretic within the previous 12 months and two other clinical/hemodynamic/echocardiographic features of HFpEF.

  • From the protocol paper we know baseline characteristics of the enrolled patients: Average age 69 years; average BMI 37; and 56% were female. Mean left ventricular EF (LVEF), 57%. Half had atrial fibrillation (AF), 80% had hypertension (HTN), one-third had coronary artery disease (CAD).

  • Follow-up is 1 year.

The dual primary endpoint is weight loss and quality of life as measured by Kansas City Cardiomyopathy Questionnaire (KCCQ), which is how patients judge their well-being, on multiple parameters. If both of those endpoints met statistical significance for superiority, then investigators would test secondary endpoints in hierarchical fashion.

First investigators calculated a win ratio of multiple outcomes, then a 6-minute walk test, then a C-reactive protein (CRP) ratio. Recall that a win ratio is an analysis wherein participant pairs are compared to see which one wins (avoids the bad outcome). The win ratio is the proportion of winners randomly assigned to semaglutide divided by the winners randomly assigned to placebo.

And here are the results: don’t act surprised.

  • Mean changes (improvements) in KCCQ were double the size in the semaglutide arm. This was highly significant.

  • Weight loss was greater in the semaglutide arm. About 13% loss of body weight in the semaglutide vs 3% in the placebo arm. That, too, was highly significant.

  • Since both were significant, the authors reported secondary endpoints.

  • 6-min-walk was better by 215 meters in the semaglutide arm. Significant again.

  • Win ratio was better for semaglutide by a whopping 1.72 — 72% better. Also significant.

  • Mean change in CRP for semaglutide vs placebo was - 43% vs -7.3% Also significant.

Comments.|Medscape Cardiology has a nice discussion on GLP-1 agonists for HFpEF. The first thing noted was how common obesity is in patients with HFpEF. Here in Kentucky, I suspect the obesity rates in HFpEF are well greater than 60%. In this discussion, they focus a lot on inflammation, and a key driver of inflammation is obesity.

So, semaglutide induces weight loss and as seen in the study results, CRP levels fall substantially. But I think explaining the benefits in STEP-HFpEF is more obvious than citing any inflammatory pathway.

  • For patients who have a mean BMI of 37, as it was in this trial, an 11% reduction in the percent of body weight is going to improve quality of life and walk time.

  • I can’t cite physiology studies, but I am nearly certain that LV diastolic parameters will also improve.

  • But HFpEF isn’t just a heart condition, it’s an all-over condition, and anything that induces that much weight loss in severely obese patients is going to help everything.

  • Again, if you use evidence to practice, this is strong evidence for benefit in patients with a serious health condition.

  • Using this drug in the longish term is not such a bad thing for people nearing their seventh decade. It’s a much different calculus when you use these drugs in adolescents or young adults who have isolated obesity.

Finally, though, I’d love to be able to induce this sort of weight loss with a lifestyle that I am immersed here in the Netherlands. You see very few obese people commuting, helmetless, on bikes.

I realize we can’t transport Dutch living everywhere, but when I wear my science hat, I wonder whether weight loss with diet, exercise, and lifestyle changes wouldn’t even exceed the benefits of GLP-1 agonists.

My worry is that these drugs may well treat a problem now,  but its success may actually worsen societal norms and move us further from the Dutch ideal.


It is said that you learn as much from trials with a non-significant primary endpoint as you do from positive trials. That is surely the case for the NOAH AFNET 6 trial, presented as the second hot line study here at ESC.

The primary endpoint of NOAH may be statistically non-significant, but the results are highly significant clinically — and scientifically.

According to the press release. The study was stopped prematurely because of futility. I wrote about the NOAH story in the preview, but I didn’t get the whole story. The data presented here are even more impressive.

Here is a very brief recap:

  • AF increases the risk of stroke.

  • There is a gradient of risk based on the presence of stroke risk factors, such as age, HTN, diabetes, HF, vascular disease.

  •  Oral anticoagulants (OAC) given to people with risk factors reduce the risk of stroke.

But these trials were done in an era when you had to have enough AF to get it documented on an ECG in a doctor’s office. Now we have devices that can capture very short duration episodes of AF. In fact, all modern dual chamber pacers and ICDs have technology that detects even minutes of AF.

Sometimes these are called atrial high-rate episodes (AHRE) because no one knows the actual duration when a super-fast atrial high rate turns into actual AF.

In their wonderfully written introduction in the rationale paper, the authors tell us the big four uncertainties of these atrial high-rate episodes.

  • We don’t know the prognostic meaning of short duration AF. Most of the observational studies that find that these episodes are associated with an increased risk of stroke also find a low absolute yearly risk of stroke. That’s critical because there is a threshold stroke risk in which bleeding from the OAC outweighs stroke risk.

  • A lot of AHREs are not precursors of AF and don’t turn into AF.

  • The timing of AHREs are not closely tied to strokes. This lack of temporality, in the language of Bradford Hill, suggests that there may be other mechanisms of cardio-embolic stroke.

  • Not all AHREs are arrhythmia; some are artifact, also known as false positives.

Here are the nuts and bolts of the trial:

  • Patients enrolled had pacers or ICDs, AHRE greater than 6 minutes, and at least one risk factor were randomly assigned to the direct acting OAC (DOAC) edoxaban or placebo.

  • The primary endpoint was a composite of stroke, systemic embolism, and cardiovascular (CV) death.

  • On the surface it was a big trial with an N of roughly 2500 patients.  But recall that new OACs vs warfarin trials enrolled more than 10,000 patients, and these were patients with clinical AF.

  • The mean age was 78 years, slightly more than one-third were women, and the median duration of the AHRE was 2.8 hours.

  • The trial was terminated early at a median follow-up of 21 months.

  • And here is what I learned that is new: It was stopped for futility of efficacy and also for safety concerns.

  • A primary endpoint occurred at a rate of 3.2% per patient year in the edoxaban group vs 4.0% in the placebo arm. Hazard ratio (HR) was 0.81 but the confidence interval (CI) was 0.60-1.08. The P-value was nonsignificant at 0.15.

  • The incidence of stroke was about 1% per patient year in both groups.

  • But safety was worse in the edoxaban arm. A safety outcome occurred at a rate of 5.9% per patient-year in the edoxaban group vs 4.5% per patient year in the placebo arm. That 31% higher rate of harm was statistically significant with a CI ranging from 1.02 (2% worse) to 1.67 (67% worse).

  • The rate of ECG-diagnosed AF in the entire cohort was 18.2% or 8.7% per patient year. Translation: Fewer than one in five patients with AHRE actually developed AF.

Comments. Oh boy! This is important data. Here we have older patients with risk factors and a median of nearly 3 hours of AHREs, which is also called subclinical AF, and  with edoxaban, no significant reduction in stroke, and a statistically significant and clinically important risk of higher rates of bleeding. That is huge.

For proponents of AF screening with any device capable of finding short-duration AF, this is like an existential punch in the gut.

  • The net benefits of AF screening were already in question after the nonsignificant findings of the LOOP trial looking at AF screening with an implantable loop recorder (ILR) in high-risk patients, which found nonsignificant stroke reduction.

  • Now, we have a direct test, a randomized controlled trial (RCT) of treating patients with short-duration AF, and it was terminated early for futility and higher rates of bleeding.

  • Proponents may say, come on Mandrola, the primary endpoint HR was 0.81, a near 20% reduction. Had the trial not been terminated it might have met significance.

  • The problem with that argument -- and I see this every day because people I work with are inclined to treat rather than not treat -- is that bleeding was increased in the edoxaban arm, and that could have also worsened with a longer trial.

  • Recall that in LOOP, the nonsignificant rates of stroke reduction were balanced by the same nonsignificant increases in bleeding.

  • Take homes: Next week in clinic, when I go home from Amsterdam, when I get a message that our patient with risk factors has “AF” on their pacer or Apple Watch: What LOOP and now NOAH teach us is that the concept of AF has changed.

Not all AF is the same. Before ESC, we had no idea of the threshold for using OAC in AF. NOAH teaches us that it’s likely greater than 3 hours, at least in patients like those in the trial.

We still have the ARTESIA trial out there, a similar trial testing apixaban vs aspirin (ASA) for subclinical AF greater than 6 minutes. The expected completion date is December of this year, and it has not been terminated yet. I do worry here about the choice of ASA as the control arm. Why? Because ASA has no benefit for stroke prevention in AF, but it will surely increase bleeding, and any rise in bleeding may mitigate the delta in bleeding between the two groups.

Renal Denervation Getting Closer

What a ride renal denervation (RDN) has had. First there was sheer glee in the early non-sham-controlled trials which found massive blood pressure (BP)-lowering effects.

Then just one sham-controlled trial, SYMPLICTY III, shattered all the joy with its essentially negative results.

Then the RDN innovators went back to the drawing board. They refined the procedure, developed two catheters – one a radiofrequency ablation catheter and the other an ultrasound catheter. They also cleaned up the trials, with more attention to standardized medical regimens.

This week, both systems — ultrasound (ReCOR) and radiofrequency SPYRAL system (by MEDTRONIC) — faced scrutiny from US Food and Drug Administration (FDA) panels. Recall that advisory panels are just that. They hear evidence from the company, FDA officials, and patients, and vote on questions. Their rulings may influence official FDA rulings, but they are just a panel.

The ultrasound system got a positive vote. The Medtronic System did not.

This is a complicated space with trials looking at BP on vs off medications.

In brief, the positive vote for the ultrasound system turned on three trials:

  • RADIANCE-HTN-SOLO: 140 patients, 2 months, mean change about 6 mmHg.

  • RADIANCE-HTN-TRIO: about 200 patients, 2 months; mean change about 4 mmHg.

  • RADIANCE-II: 150 patients, 2 months, mean change about 6 mmHg.

  • The panelists were unanimous on safety, 8 to 3 on efficacy, and 10 to 2 on benefits outweighing risks.

The big problem though, and all agree, is that we don’t know about durability of effect. Another issue is that the company is asking that the device be used in patients who inadequately respond to or are intolerant of BP medications. But this isn’t what was studied in the RADIANCE trials.

This issue reminds me of the left atrial appendage occlusion (LAAO) issue. That device was approved for patients who were excluded from the seminal trials.

A day later the FDA panel came to different conclusions on the SPYRAL catheter from Medtronic. Here there two influential trials.

  • SPYRAL HTN OFF (off medication): About 320 patients; about 4 mmHg drop in BP at 3 months.

  • SPYRAL HTN ON (on medication): Did not meet efficacy endpoint of difference in BP.

  • The FDA panel was again unanimous on safety, 7 to 6 for efficacy but 6 to 7 against on benefits vs risks.

Comments.  I don’t know what FDA will do, but I worry about this procedure. My main issue is with efficacy — these are modest reductions in BP —  and we have no idea on long-term durability.

If FDA approves, then I will have to say that I don’t understand regulation. Here is why. The main question is whether this modest effect lasts more than months, and whether it reduces outcomes.

All the regulatory bodies have to do is force the companies to run longer sham-controlled trials. The nanosecond you break blinding, you lose any ability to infer effect size.

A paper published last April in the Lancet purported to show long-term effects out to 3 years, but that is an incorrect conclusion, because in this analysis, blinding was terminated at 6 months.

  • I have tension here. This is an interesting technology and the investigators deserve credit for iterating and improving on the early techniques.

  • If it worked long-term, that would be a big thing, because medicine and lifestyle modification are hard to comply with.

  • But if you think expensive, likely low value procedures like LAAO and pulmonary artery monitoring are overused, RDN could dwarf them all, because HTN is so prevalent.

  • If the labeling is for patients who do not tolerate BP meds and re-imbursement from payers is generous, RDN could break the healthcare budget.

The simple solution is right before our eyes: Just one longer trial maintaining blinding.

Maintenance of Certification Debate

In the United States, there has been a rekindling of a debate regarding certification of physicians.

I realize this podcast has many international listeners, and this is a decidedly American debate, so I will keep my comments brief, and also try to hit big themes rather than specifics. I suspect this debate pales in comparison to what British colleagues are dealing with in the National Health Service.

The gist of the maintenance of certification (MOC) issue is that the American Board of Internal Medicine (ABIM) has changed the rules of continuing to be called a “board certified” physician. This is more than a label; being board certified generally equates to being able to work as physician.

When I started in medicine, all you needed to do was pass a test and you were good for life, as long as you did the CME and kept up your privileges.

The next step the ABIM took was to say, nope, medicine is changing too fast, now you need to pass another big, high-stakes, expensive test every 10 years. (I’ve passed electrophysiology boards three times; 2026 will be number 4, if the world and I last until 2026)

But now, it’s even worse. Now, the ABIM requires you to take and pay for something called longitudinal knowledge assessments (LKA) every year.  These are open book tests that some say they test how fast you can you use Google. For instance, a person could finish a residency and fellowship and pass the big test, but if he or she does not pay to take the LKA the very next year, they are no longer certified. This is nuts because the most up-to-date doctors in the world are those just out of fellowship and fresh off passing their first board exam.

The big-picture specific issues are (and I will surely miss some):

  • ABIM is an unaccountable organization.

  • ABIM has relationships with other healthcare organizations and professional societies.

  • ABIM isn’t exactly transparent with their finances.

  • ABIM has ensured a monopoly because boards of licensure, hospital organizations, and payers require it.

  • There exists no evidence that ABIM-branded medical education is better than standard CME. 

I will note some general themes.

  • The debate is occurring on social media, particularly in the realm of podcasts. The debate must occur outside the medical societies and journals, because many of our societies make money selling test prep. It is a classic conflict of interest.

  • My friend, oncologist, Chadi Nabhan, hosted a debate featuring practicing hematologist Aaron Goodman and the CEO of ABIM Richard Baron on his Healthcare Unfiltered podcast. Another friend, Wes Fisher, a practicing electrophysiologist, has been in the forefront representing doctors, using his blog.

  • Longform is the first big point. I may be wrong, but I think platforms that allow for longform debate are a hugely positive thing for medicine and society. The most popular podcast in the world routinely has 2- to 3-hour long discussions. Moving away from the 2- to 4-minute sound bytes and rapid fire presentations is a positive.

  • Another global point to make here is that the reason American physicians will not win against unaccountable rent-seekers is because we are not organized.

  • Stop and think about it at a basic hospital level. When you walk into your hospital, have you ever thought that all this — I mean all of it — exists because clinicians see patients and send bills to payers for their work.

  • If not for the working stiff clinicians, none of it exists. If we were organized, we would have influence.

But I am pessimistic that we could ever organize. When I began cardiology private practice, it was a constant struggle to keep 20 cardiologists together in a group. We were eventually broken up when hospitals started employing doctors.

So, my message is that until we in the profession support each other and organize together, we will be susceptible to arbitrary rules and rent-seeking unaccountable organizations.

I will watch how this profession transforms itself over the coming years, but if we are to remain a profession, we must find a way to organize. Professional societies, I am afraid, are not the answer.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.